Graph-based timing analysis timing calibration
Abstract
A set of multi-corner multimode (MCMM) databases that correspond to a set of working scenarios are accessed. A full timing update on the set of MCMM databases, for the set of working scenarios, is applied. A graph based analysis (GBA) timing calibration is performed on the databases, for the set of working scenarios to obtain a set of GBA-calibrated databases. Multiphase optimizations on the set of GBA-calibrated databases are iteratively performed to generate a set of optimized databases, including: performing a phase-specific optimization on the set of GBA-calibrated database to obtain an improved set of databases, and recalibrating GBA timing on the set of improved databases prior to a next phase-specific optimization.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1. A system, comprising:
a processor configured to:
access a set of multi-corner multimode (MCMM) databases that correspond to a set of working scenarios;
apply a full timing update on the set of MCMM databases, for the set of working scenarios;
perform a graph based analysis (GBA) timing calibration on the databases, for the set of working scenarios to obtain a set of GBA-calibrated databases; and
iteratively perform multiphase optimizations on the set of GBA-calibrated databases to generate a set of optimized databases, including:
performing a phase-specific optimization on the set of GBA-calibrated databases to obtain an improved set of databases; and
recalibrating GBA timing on the set of improved databases prior to a next is phase-specific optimization; and
a memory coupled to the processor and configured to provide the processor with instructions.
2. The system of claim 1 , wherein a MCMM database of the set of MCMM databases comprises a graph representation of a circuit design.
3. The system of claim 1 , wherein a MCMM database of the set of MCMM databases comprises a graph representation of a circuit design and timing arc information in the graph representation.
4. The system of claim 1 , wherein each MCMM database of the set of MCMM databases comprise a netlist.
5. The system of claim 1 , wherein the full timing update includes path based analysis (PBA) and GBA.
6. The system of claim 1 , wherein the processor is further configured to apply calibration to each of the set of MCMM databases for each of the set of working scenarios.
7. The system of claim 1 , wherein the processor is further configured to release a PBA database for a plurality of working scenarios in the set of working scenarios.
8. The system of claim 1 , wherein the phase-specific optimization includes a PBA and a GBA.
9. The system of claim 1 , wherein the processor is further configured to release a PBA database after the step of recalibrating GBA timing.
10. The system of claim 1 , wherein recalibrating GBA timing comprises:
for an input pin in a PBA database, in the event a PBA path exists on the input pin:
determining a most critical PBA path on the input pin; and
calculating an arrival time offset on the input pin; and
updating GBA timing with the calculated arrival time offset.
11. The system of claim 10 , wherein calculating the arrival time offset is based at least in part on a difference between PBA and GBA arrival times.
12. The system of claim 10 , wherein calculating the arrival time offset is based at least in part on a difference between a difference of PBA and GBA arrival times.
13. A method, comprising:
accessing a set of multi-corner multimode (MCMM) databases that correspond to a set of working scenarios;
applying a full timing update on the set of MCMM databases, for the set of working scenarios;
performing a graph based analysis (GBA) timing calibration on the databases, for the set of working scenarios to obtain a set of GBA-calibrated databases; and
iteratively performing multiphase optimizations on the set of GBA-calibrated databases to generate a set of optimized databases, including:
performing a phase-specific optimization on the set of GBA-calibrated databases to obtain an improved set of databases; and
recalibrating GBA timing on the set of improved databases prior to a next phase-specific optimization.
14. The method of claim 13 , wherein the full timing update includes path based analysis (PBA) and GBA.
15. The method of claim 13 , further comprising applying calibration to each of the set of MCMM databases for each of the set of working scenarios.
16. The method of claim 13 , further comprising releasing a PBA database after the step of recalibrating GBA timing.
17. The method of claim 13 , wherein recalibrating GBA timing comprises:
for an input pin in a PBA database, in the event a PBA path exists on the input pin:
determining a most critical PBA path on the input pin; and
calculating an arrival time offset on the input pin; and
updating GBA timing with the calculated arrival time offset.
18. The method of claim 17 , wherein calculating the arrival time offset is based at least in part on a difference between PBA and GBA arrival times.
19. The method of claim 17 , wherein calculating the arrival time offset is based at least in part on a difference between a difference of PBA and GBA arrival times.
20. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
accessing a set of multi-corner multimode (MCMM) databases that correspond to a set of working scenarios;
applying a full timing update on the set of MCMM databases, for the set of working scenarios;
performing a graph based analysis (GBA) timing calibration on the databases, for the set of working scenarios to obtain a set of GBA-calibrated databases; and
iteratively performing multiphase optimizations on the set of GBA-calibrated databases to generate a set of optimized databases, including:
performing a phase-specific optimization on the set of GBA-calibrated databases to obtain an improved set of databases; and
recalibrating GBA timing on the set of improved databases prior to a next phase-specific optimization.Cited by (0)
No later patents cite this yet.
References (0)
No backward citations on record.